Congenital adrenal hyperplasia (CAH) is an inherited inability to synthesize cortisol. More than 90% of cases are caused by deficiency of steroid 21-hydroxylase (CYP21), which occurs in the severe classic form in ~1:16,000 births. Infants with classic CAH are susceptible to life threatening adrenal insufficiency. Additionally, the adrenal cortex overproduces cortisol precursors, particularly 17-hydroxyprogesterone (17-OHP), which are further metabolized to androgen precursors (e.g., androstenedione) and subsequently to testosterone. Severely affected girls are born with ambiguous external genitalia. Inadequately treated patients are exposed to high levels of sex hormones which promote rapid somatic growth and accelerated skeletal maturation leading to premature epiphyseal fusion. Adult heights in classic patients average ~ 7 cm below the population mean. Patients are treated by replacing glucocorticoids (usually with hydrocortisone) and mineralocorticoids (with fludrocortisone). Both under-treatment and over-treatment with glucocorticoids put patients at risk for short stature, the former owing to premature epiphyseal closure induced by abnormal secretion of sex steroids, and the latter to glucocorticoid-induced inhibition of growth. Excess glucocorticoid exposure also puts CAH patients at risk for excess weight gain and metabolic syndrome. This project will test whether pharmacologic blockade of sex steroid synthesis in prepubescent children with CAH improves these problems. The study drug, abiraterone acetate, is a prodrug of abiraterone, an androgen biosynthesis inhibitor approved for treatment of prostate cancer. The study population includes prepubescent children with classic CAH owing to 21-hydroxylase deficiency, girls 2-8 and boys 2-9 years old. Subjects will be treated with hydrocortisone and fludrocortisone throughout each study. In a Phase 1 study, we will determine the minimum effective dose of abiraterone acetate that normalizes androstenedione levels. During the 7-day Treatment Period, up to 3 cohorts of 8 subjects each will receive, in succession, 1, 2 or 4 mg/kg/d of abiraterone acetate, until 7/8 subjects have normalized their morning androstenedione concentration on Day 7 of abiraterone administration. In a Phase 2 study, we will assess the utility of abiraterone acetate as adjunctive therapy to minimize excessive androgen secretion and allow more physiologic glucocorticoid replacement. This study is a double-blind randomized controlled trial of abiraterone acetate (dose determined in Phase 1) versus placebo for 24 months. We will randomize 54 subjects (36 to abiraterone and 18 to placebo), aiming to have 48 subjects completing the study. The primary hypothesis is that subjects receiving abiraterone will have slower skeletal maturation (i.e., advancement in radiologic bone age). The secondary hypothesis is that subjects who receive abiraterone will require lower mean daily hydrocortisone doses to normalize androstenedione. Exploratory hypotheses are that subjects receiving abiraterone will have lower increases in weight, height and BMI, and consequently less insulin resistance as measured by HOMA-IR, than subjects receiving placebo.